Structured low-rank reconstruction for navigator-free water/fat separated multi-shot diffusion-weighted EPI

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Medical Physics
• Image Reconstruction

Background:

• Multi-shot echo-planar imaging (EPI) enhances MRI resolution but is susceptible to motion-induced phase variations.
• Combining EPI with chemical-shift encoding (Dixon) separates water/fat signals but exacerbates motion artifacts.
• Existing methods often require navigators or struggle with simultaneous motion correction and water/fat separation.

Purpose of the Study:

• To develop a navigator-free algorithm for simultaneous water/fat separation and motion correction in multi-shot diffusion-weighted EPI.
• To address the challenge of shot-to-shot phase variations in hybrid EPI-Dixon MRI sequences.
• To improve the robustness and applicability of diffusion-weighted MRI.

Main Methods:

• An iterative, model-based reconstruction pipeline utilizing structured low-rank regularization.
• Data-driven estimation and elimination of shot-to-shot phase variations.
• Simultaneous separation of water and fat images using chemical-shift encoding.

Main Results:

• The proposed algorithm effectively corrects phase variations and separates water/fat signals in simulations and in-vivo experiments.
• Demonstrated superior performance compared to extra-navigated and self-navigation approaches.
• Enabled reconstruction of under-sampled datasets and facilitated partial Fourier reconstruction.

Conclusions:

• The developed algorithm offers a robust solution for eliminating phase variations and separating water/fat images.
• It shows significant promise for advancing diffusion-weighted MRI across diverse anatomical regions.
• The navigator-free approach simplifies scan protocols and enhances clinical applicability.